When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
I'm still having a long cold start issue with my Bosch IIIs. In a previous thread I verified that the oil pressure switch is fine, the fuel pressure is fine, the fuel pump primes, and I have good spark. I'm now leaning the direction of tuning since everything else looks to be OK. I set the Injector Offset vs Battery Voltage table to values for my injectors and zero'd out the Injector BPW vs Reference Pulse table (hopefully those are close to the right names, don't have my tuning computer with me at the moment). It doesn't seem to have helped the cold start too much though.
In the morning, I have to crank it for several seconds, where I can hear it start to catch, but it never really does, so I let off for a second, and crank it again. It usually fires up much easier the second time after letting it rest a second.
So my question is what table should I attack next? I see there's several related to cranking, what would be the most likely culprit? I'm running AUJP.
When you say fuel pressure is "fine", does it actually hold pressure in the rail after priming, or does the pressure slowly dissipate...?
Reason I ask is a while ago I bought an adjustable regulator that had an internal leak. It would bleed off pressure back to the fuel tank and the rails wouldn't hold pressure during cranking, such that it behaved similar to what you're describing.
I believe it will slowly bleed down, however I've checked the pressure gauge while cranking and it holds for the duration of the cranking where it should. I've even tried jumping the fuel pump relay while cranking and it didn't help any.
Mostly stock. Only engine mod is the injectors and headers. The chip is an AUJP and the only other modification I've done is with the spark advance when cold to not use the highest table values.
what are your BLMs like when it's running? If they're way off, it could be indicative of an issue with how you've specified injector size in the calibration.
It's had this problem since I was all stock, I did the spark advance change after the injectors, when I had my 2600 stall converter in. It was causing a sudden change in power when the car reached a certain temperature. So the stock tune didn't work either, even without my spark advance change.
You need to change the "Crank PW multiplier vs. Reference Pulse" table. Add some fuel into the first 8 table entries. I have mine at 0.5 for the first 8. I ended up using stock voltage offsets on my Bosch III 24's. It seemed to work better that way overall.
I'm running $6E on a MAF car. But it shouldn't matter much for startup. The MAF is pretty much ignored during start.
That change certainly helped, but it doesn't quite get there yet. Should I try increasing them to 0.8 or something and give it another shot? Also potentially interesting is that it usually starts up better in very cold weather than warm weather. Although there's no cold weather recently to test with .
I used the ARAP table, which for a reason I have yet to understand has the first 8 entries zeroed out. I put them at .5 I think entry 9 is also .5 and then it goes to .25 and tapers off to .20. The rest of the table I left as-is. I have considered upping the other entries and keeping the fueling going at .5 for longer. You may have to play with it for sure.
Alright, I'll have to change it around a bit and see what I can do. Only one maybe two tests per day is unfortunate though. For AUJP it's a really weird table, it's 1.0 for the first two, then 0's for a couple entries, then 0.8, then 0's again, then 1.0 for a couple. Can't say I quite understand the logic behind it all
I don't understand the logic behind it either. The ARAP one has nothing for the first 8 entries then tapers off slowly. The other ones I've looked at were all over the place too. Going with ARAP and the first 8 set to .5 gets me 90% of the way there I would say. I too am in the same boat with tuning it. I get one cold start per day and the rest of the time it starts normally. I bet that's why it was all over the place on the various factory tunes. Same situation. My visor says to hold the throttle open if it doesn't start first try in cold weather..... my guess is the factory maps weren't perfect in all conditions on all models either.
When I got home I set the first eight entries to 0.8. When trying to do a hot start however it flares up, then drops for a bit, then continues. So I decided to instead try 1.0 for the first couple, then a couple 0.8s, then some 0.5s, then finish it off with 0.3 and 0.2. Just guessing at this point but I'll just have to keep adjusting until both the cold start and hot start work well. If anyone has some wisdom on what to put in the table I'm all ears.
Well, we're certainly making progress. This morning it caught hard pretty quick after cranking but it died out right after that. Not sure if I need to add fuel in the start, pull fuel after the start, or add fuel later in the table. I assume that at some point it will stop using the ref pulse table if it detects that the engine has started? Maybe I should look in the code so that I understand what it's doing during cranking better.
An overlooked possibility is a build up of carbon on the backs of the intake valves. The Bosch III's spray a finer spray then the OEM injectors. The carbon absorbs the fuel until it gets saturated.
Then the engine finally starts.
As for the best values for the DRP versus cranking fuel table. GM was all over the place, check various BINs and see what was done. Be sure to also look at Y-body BINs such as AXCN and ANHT.
The cranking PW code is rather straight forward, easy to go through.
Shouldn't be the carbon issue as this is a reman engine and the injectors were installed at the same time as the engine went in. So had the problem since it was first installed. It is interesting that adding fuel makes the engine cut out after it starts (at least hot). I assume this is why the original tables only had two 1.0 cells then a bunch of zero cells. I can only assume that once it starts catching it dumps fuel in for some reason. Tracking down some other bins is a good idea.
Hmm, don't see those in my list of tables. Maybe it's just named something different for the xdf I'm using? If I have those tables I'd definitely want to look at them.
The car certainly wasn't too happy with the way I set the table up; today when I started it after work it was particularly finicky. It caught and immediately died, then on the next attempt or two it started, but would rev up a bit (< 800 RPM I'd say) and dip down low, then repeat.
I did gain some insights from looking at the code. First off, the 'Crank Fuel Delivery Delay vs. MAT' table does exactly what it implies, it delays firing injectors for however many distributor reference pulses based on temperature. It does not start looking at the 'Crank Pulse Mult. Vs. Ref Pulses 1-24' table until after that. Once it does get past the delay it starts at the first cell of the PW vs Ref Pulse 1-24 table. I don't think I should really have any problems with the 'Crank Fuel Delivery Delay vs. MAT' table, all it will do is increase the initial time before it will try firing.
I also read that the distributor reference pulse occurs four times per crank revolution, which corresponds to a firing event per reference pulse. That means that when I was increasing the first 8 or so cells of the PW vs Ref Pulse table, it would only be for the first 2 revolutions before it would go back to the original table values. I also found that after the first 16 reference pulses it will continually loop over 17-24 if you keep cranking it.
Looking up cranking information, I came across how MegaSquirt does it. It looks like they batch fire once per ignition event (so distributor reference pulse). I guess there's also a concept of firing an initial shot to wet the intake. So I'm considering trying an initial couple 1.0 values like it had stock, then fill in either every reference pulse or every other pulse with a value. I just have to be careful not to flood doing this. The MegaSquirt page also suggested that if you open the throttle a bit and it gets easier to start that I need to pull fueling. I might end up needing to mess with the Crank Pulse Width Vs Temperature table to make sure it fires well at various temperatures after doing this.
I'm having excellent results on cold start with a smoothing of the Crank PW Multiplier vs. Ref Pulse like so:
Warm restart is a bit spotty now. I've had to open the throttle a bit so it seems most likely rich. I'm going to switch to the ARAP Crank PW vs. Coolant Temp table which has about 30% lower PW values for the 111 to 241 degree blocks. Being I'm running 24 lb injectors that are scaled at 23, I'm much closer to the 350 injector sizing than I am the 305's 19 lb injectors.... not sure if the injector size constant comes into play with cranking or not.
Pretty sure I'm not dealing with any carbon on my intake valves either. That's usually not much of a problem on port injected engines but I've also had my manifold off in the last few thousand miles and saw no buildup. I've replaced the stem seals recently also.
Now that the cold start is working well with this table, fixing the warm restart seems much easier since I can get it to do that more than once or twice a day.
GD
Last edited by GeneralDisorder; 07-07-2017 at 10:19 PM.
I may end up going to a taper like that. Right now I'm going to try 1.0 for the first two cells followed by 0.3 for everything else. I should be able to test a (mostly) cold start and hot start tonight and see if it works alright. If only I had 10 engines and a climate controlled room to test them in
Since you already looked at the code - does the cranking code take the injector size into account? The thing I noticed is the ARAP (350) code has lower PW vs Coolant Temp than the 305 code on the hot side. It has similar or longer PW on the cold side but an aluminum head would probably need more fuel on cold start due to creating more fuel condensation in the port and chamber from more rapid heat transfer vs cast iron..... well that's my speculation anyway.
I don't recall seeing that but I didn't fully trace the code (it's in many places). Maybe someone with more experience with the code could chime in on that one.
The reduction in Crank PW vs. Coolant Temp (111 to 241) from around 11 to 7.81 made a big difference. Warm and hot restart are working much better. I'll report back after more testing but other than possibly a further reduction to refine it I don't think it will need much tweaking.
Since you already looked at the code - does the cranking code take the injector size into account?
GD
Cranking fuel doesn't use the injector flow rate parameter. It uses a single scalar as the maximum PW, this is the parameter to change for different flow rate injectors and/or engine displacement:
Code:
For $6E
LC371 FDB 13108 ; 200.01, SCALAR FOR MAX PW
; CAL = SCALAR * 65.536
For $8D
L83AD: FDB 13108 ; SCALAR FOR MAX PW
The cranking PW tables modify this value by reducing it along the way.
I knew something must have been up when the 350 code for 22lb injectors was wildly different than the 305 code for 19lb. But besides being wildly different they also aren't any kind of consistently different so it's hard to tell what GM was using as criteria for programming these tables.
That makes a lot of sense now why you have to mess with all these crank tables when switching injector sizes.
I added LC371 to my XDF and checked through a dozen or so BIN's. It's always 200 msec. So now it makes sense why all the changes to various BIN's. GM didn't use calculations based on physical engine and injector characteristics - it's all done by dead reckoning from a basis of 200 msec pulse width.
With Southbay 24's on my LB9, These settings are working well for me:
Changing the table to all 0.3's except the first two (1.0) really helped. I can start it up on the first crank now at least, and it's fairly happy with hot starts. There was still a bit of a stumble the first time it wanted to catch so maybe I'll try upping the 3rd and 4th cells a bit. Unfortunately I don't have much weather variation right now so tuning the PW vs temp table is going to have to wait.
Weather variation doesn't matter much really. The engine temp delta is much higher than your typical weather delta. Right now it's about 55 at night, and maybe 90 durring the heat of the day. So a delta of 35 degrees F. Engine temp runs about 200 F give or take. That means a delta of 145 F in the morning, or 110 F in the middle of the day..... its still a LOT of Delta.
Sure, it's hard to judge startup sub-freezing conditions at the moment, but I think my tables are sufficiently rich that it won't matter. Time will tell.
I added LC371 to my XDF and checked through a dozen or so BIN's. It's always 200 msec. So now it makes sense why all the changes to various BIN's. GM didn't use calculations based on physical engine and injector characteristics - it's all done by dead reckoning from a basis of 200 msec pulse width.
With Southbay 24's on my LB9, These settings are working well for me:
GD
As for the 200 msec MAX PW, it is the same for an LB9 and L98 due to the ratio of injector flow to engine displacement. For example, divide the displacement by the injector flow, it comes out to 16 for both the LB9 and L98.
So now install 24 #/hr injectors on an LB9 (305), and that ratio is much lower (305 / 24 = 12.7). So it is going to be richer during cranking. To change that need to change the crank scalar for MAX PW. Then the remaining tables fall into place as-is (for the most part).
Taking 19 #/hr and divide it by 24 #/hr for a result of 0.79, so multiply the MAX PW scalar by .79 to make it a MAX PW of 158.
Ah! I hadn't considered that the ratio of injector to displacement would yield the same factor..... interesting. Thank you for that insight.
I hadn't found this scalar discussed anywhere in my searches on the subject. It's not defined in any of my XDF's (till now) - which I specifically searched out the most expanded $6E def I could locate....
For my application at least, it seems to want more fuel (or more fuel earlier anyway) on cold starts and less on warm/hot starts. So changing the scalar alone doesn't seem like it would do that job. May have something to do with my 1.6 rockers. I'm sure this is causing at least a small amount more air. The hot restart falls right in line with injector sizing since it's obviously rich there.
Still don't understand why all the weird mapping of the PW vs Ref Pulse table on different BIN's. Even stranger considering the same MAX PW scalar.
GD
Last edited by GeneralDisorder; 07-09-2017 at 12:17 PM.
Your engine is modified, so I would expect some tweaking of the crank fueling parameters (other then the MAX PW scalar). There are other areas that affect cranking, such as the speed at which the starter cranks the engine over at.
The IAC steps during cranking, how much more cam is there. A cam with more duration and overlap is not going to create as much vacuum as a stock cam. This means a lower airflow and more reversion, which means it is more difficult to pull the fuel into the chamber.
A hot engine will likely crank at a higher RPM.
Don't forget the base distributor setting. That is the SA the engine uses during cranking.
And the load difference between a stick car and an automatic equipped car.
> Still don't understand why all the weird mapping of the PW
> vs Ref Pulse table on different BIN's.
Different tuners took different approaches. The main thing is to not flood the engine, but at the same time get it to kick off quickly & reliably (ARAP aside, but it is reliable).
Both of the following inject the same mass of fuel (for the 8 DRPs shown):
0 1 1 0 0 0 0 0
.25 .25 .25 .25 .25 .25 .25 .25
But there will be a difference in how the engine cranks, catches and transitions to running fuel.
Very insightful! I appreciate all that information and I'm sure others will as well as this will surely show up in searches.
I do have a mini-starter so that's going to affect the cranking - they seem to ramp up slower but spin faster than the old school non reduction starters.
The lesson here kids, is that even minor changes like newer style (but same wide open flow rate) injectors, some 1.6 rockers, and a gear reduction mini-starter can dramatically affect how the fuel injection system operates.
Either stick to 100% stock components for everything, or get comfortable with tuning your own car. Not many professional tuners doing these anymore and would still cost a lot of money to do it incrementally. Things like cold startup is going to be very hard for even an experienced tuner to dial in with a single tuning session, and there's no good way to datalog some of this stuff - during cranking the ECM data stream is not fast enough and some of the critical data items like reference pulse count are not available unless you dive into the code.... which is unlikely to be something you would want to pay someone to do if you are paying for tuning services.
RBob - you are amazing. And I'm going to get your EBL system and 7730 swap my car prior to the Pro-Charged 383 going in. With your support it makes more sense than any other option.
I added LC371 to my XDF and checked through a dozen or so BIN's. It's always 200 msec. So now it makes sense why all the changes to various BIN's. GM didn't use calculations based on physical engine and injector characteristics - it's all done by dead reckoning from a basis of 200 msec pulse width.
GD
Hi GD, just read this old post, and I want to ask you how you added this scalar to your XDF? like where is location LC371 in the XDF? I'm using TP RT, latest version, and the 6E expanded with CCP.xdf. I been editing and modifying things, in this xdf, but really need to know how add/create new tables and scalars.....
thanks,
Cranking fuel doesn't use the injector flow rate parameter. It uses a single scalar as the maximum PW, this is the parameter to change for different flow rate injectors and/or engine displacement:
Code:
For $6E
LC371 FDB 13108 ; 200.01, SCALAR FOR MAX PW
; CAL = SCALAR * 65.536
For $8D
L83AD: FDB 13108 ; SCALAR FOR MAX PW
The cranking PW tables modify this value by reducing it along the way.
RBob.
Ok RBob, I have to question you on this stuff, if you dont mind. I understand about the max PW scalar, and it being at 200mS. What i dont get is how this value is figured in with the crank fuel PW calculation. If you look at GD tables above, at 68degF his PW is 15.63mS, and with a multiplier of 0.5 for the first few DRPs, gives you a crank fuel PW of 7.815mS. Nothing too complicated, so how does the 200mS max PW value get added in?
And/also, i was able find this location 371 in my xdf. The value there is 33hex or 051dec. I figured out how to add this scalar into my xdf, but your conversion listed doesnt make sense to me. There must be some other conversion to get the value displayed equal to 200.01 mS.
And lastly, looking at your other post, referring to ratio-ing this value based on injector size and engine size. i think i get it, but is modifying this 200mS scalar, really necessary or important when making injector changes? I'm running a 350TPI, with stock 22 lb injectors, but eventually I will be going with 24lb B3 units come springtime. So, if my math is correct, i would need to change my max PW scalar to 183.3 once i goto 24lb injectors. Agree?
If the OP's cold start issue has not be solved, have you tried to start cranking while the pump is priming, say waiting 1 second and then beginning to crank. This will ensure that you are cranking with maximum pressure to the injectors and fuel should be at the injectors. If there is a pressure bleed-off, cranking after pump prime ends can cause what you are seeing. Ditto when hot. Also, given you went from (I assume Multec's) to B3s, you really should not require any major changes for cranking. Just zero the low PW table as you did.
If you look at GD tables above, at 68degF his PW is 15.63mS, and with a multiplier of 0.5 for the first few DRPs, gives you a crank fuel PW of 7.815mS. Nothing too complicated, so how does the 200mS max PW value get added in?
Bottom Line: Changing the 200ms PW Scalar by X% will correspondingly change Cranking PW by X% at ALL Temps.
If you're interested, the attachment describes exactly how cranking PW is calculated and how the fuel is sent the injectors using the multiplier tables you referenced. Granted this is for $8d, but other masks are not materially different.
Originally Posted by MikeT 88IROC350
And/also, i was able find this location 371 in my xdf. The value there is 33hex or 051dec. I figured out how to add this scalar into my xdf, but your conversion listed doesnt make sense to me. There must be some other conversion to get the value displayed equal to 200.01 mS.
The Scalar is a 2-byte value at 0x371/372 and the conversion is: X/ 65.536. An entry of 200 results in 0x3334 = 13108 / 65.536 = 200....
Sorry for the 2 files, but unable to delete one. Both are the same except for title.
Last edited by 84Elky; 01-11-2020 at 01:31 AM.
Reason: Tried to delete file
And lastly, looking at your other post, referring to ratio-ing this value based on injector size and engine size. i think i get it, but is modifying this 200mS scalar, really necessary or important when making injector changes? I'm running a 350TPI, with stock 22 lb injectors, but eventually I will be going with 24lb B3 units come springtime. So, if my math is correct, i would need to change my max PW scalar to 183.3 once i goto 24lb injectors. Agree?
Elky answered your other questions. As for the above, I agree, 183.3 should be correct.
Bottom Line: Changing the 200ms PW Scalar by X% will correspondingly change Cranking PW by X% at ALL Temps.
If you're interested, the attachment describes exactly how cranking PW is calculated and how the fuel is sent the injectors using the multiplier tables you referenced. Granted this is for $8d, but other masks are not materially different.
The Scalar is a 2-byte value at 0x371/372 and the conversion is: X/ 65.536. An entry of 200 results in 0x3334 = 13108 / 65.536 = 200..
Huh. So is this 200ms PW scalar ONLY used during cranking?? And a 2-byte value, didnt know that. So i tried it, put the conversion in and the displayed value goes to 0. The 2 locations in the hex code goto FF. Not sure WTF i am doing now. Do you have to put the same conversion in at location 372? i had 33 in at 371, and 34 at 372, so i know it should come out right, but now they are stuck on FF.
Also, there are 2 other scalars, one for min IBPW, and one for default IBPW. Both of mine are at 1.59ms. Both of these have a conversion of X/65.536, and values in hex are 68 or 104 dec; this math seems to check out. So i been reading up on what these do. Seems kind of confusing, why need both? The default one is used if the calculated PW is <= the value. If the min IBPW is less than that, it will never get to it, based on the default!?
So, in the crank fuel PW table, no sense having a value less than 1.59, cuz it will never go that low?
Lastly, downloaded your 2 files. Gonna take me while to figure that all out. I wish it applied to the $6E definition, but i do have a C4 that uses $8D, so i can roll with that. Another day!
Yes, the crank PW scalar is only used during cranking. The TunerPro entry for $6E should look like this:
RBob.
Ok, Rbob, thanks. I did get it figured out, before i saw your post. Went back to a previous bin, deleted the scalar entry, and started over. The 2 byte size thing got me originally, but all good now. So the only time this value should be changed, is if your engine size changes, or injector size changes? And of course you can change the other crank fuel tables, as needed......i will have to go into my $8D definition, and pull out this scalar, since i have the 24 B3s in my '90 C4.
So the only time this value should be changed, is if your engine size changes, or injector size changes? And of course you can change the other crank fuel tables, as needed.....
It is a global change that affects how much fuel (PW) will be available while cranking. Again, if changed +/- X%, available cranking fuel will change by +/- X%. Can similarly change the available fuel at individual temps in the table below the scalar. But note that this PW will be factored and provided according to the settings in the cranking fuel delay tables. See my attachment in above post.
07-05-2017, 11:36 AM
. 
